Tribo-performance analysis of red mud filled glass-epoxy …dspace.nitrkl.ac.in/dspace/bitstream/2080/791/1... · 2010. 4. 12. · Taguchi’s parameter design on the erosion wear

Tribo-performance analysis of red mud filled glass-epoxy composites using Taguchi

experimental design Sandhyarani Biswas and Alok Satapathy

aDepartment of Mechanical Engineering, National Institute of Technology, Rourkela 769 008, Orissa, India

Accepted in Materials and Design (2009)

http://dx.doi.org/10.1016/j.matdes.2009.01.018

Archived in Dspace@nitr

http://dspace.nitrkl.ac.in/dsapce

Tribo-performance analysis of red mud filled glass-epoxy composites using

Taguchi experimental design Sandhyarani Biswas and Alok Satapathy

Department of Mechanical Engineering National Institute of Technology, Rourkela 769008, India

ABSTRACT

Solid particle erosion of polymer composites is a complex surface damage process, strongly affected by material properties and operational conditions. To avoid repeated experimentation, it is important to develop predictive equations to assess material loss due to erosion under any impact conditions. This paper presents the development of a mathematical model for estimating erosion damage caused by solid particle impact on red mud filled glass fiber reinforced epoxy matrix composites and also a correlation derived from the results of Taguchi experimental design. Red mud is an industrial waste generated during the production of alumina by Bayer’s process. Using this red mud as the filler, hybrid glass-epoxy composites are prepared and experiments are conducted to study the erosion wear behaviour of these composites and the results are compared with the predicted values. Using Taguchi method for analysis, the significant control factors and their interactions influencing the wear rate predominantly are identified. The filler content in the composites, erodent temperature, the impingement angle and velocity are found to have substantial influence in determining the rate of material loss from the composite surface due to erosion. Key Words: Hybrid Composites, Epoxy, Red Mud, Erosion, Modelling, Taguchi method

1. Introduction

Polymers find wide engineering applications due to their low density, reasonably

good strength and wear resistance as compared to monolithic metal alloys. For weight

sensitive uses, undoubtedly they are the most suitable materials but prohibitive costs and

stability of properties pose challenge for the researches in the process of development of

composites. In order to bring down the cost, cheap and easily available fillers are a viable

option. However, mechanical properties of the composites should not be degraded in the

attempt of reducing the cost. Therefore, purpose of using fillers is twofold: first, to improve

the mechanical, thermal or tribological properties, and second, to reduce the cost of the

component. Specifically, in polymers, a large number of materials such as minerals and

inorganic oxides (alumina and silica) are mixed with thermoplastics like polypropylene and

polyethylene [1]. Through judicious control of reinforcing solid particulate phase, selection

of matrix and suitable processing technique, composites can be prepared to tailor the

properties needed for any specific application. In past two decades, ceramic filled polymer

composites have emerged as a subject of extensive research. But due to high cost of

conventional ceramic fillers, it has become important to explore the potential of cheap

materials like mineral ores and industrial wastes for utilization in preparing particle-

reinforced polymer composites.

Production of alumina from bauxite by the Bayer’s process is associated with the

generation of red mud as the major waste material in alumina industries worldwide.

Depending upon the quality of bauxite, the quantity of red mud generated varies from 55-

65% of the bauxite processed [2]. The enormous quantity of red mud discharged by these

industries poses an environmental and economical problem. The treatment and disposal of

this residue is a major operation in any alumina plant. Red mud, as the name suggests, is

brick red in colour and slimy having average particle size of about 80 μm. It comprises of the

iron, titanium and the silica part of the parent ore along with other minor constituents. It is

alkaline, thixotropic and possesses high surface area in the range of 13-16 m2/g with a true

density of 3.30g/cc. Depending on the source, these residues have a wide range of

composition: Fe2O3 20–60%, Al2O3 10–30%, SiO2 2–20%, Na2O 2–10%, CaO 2–8%, TiO2

traces 2–8%. The leaching chemistry of bauxite suggests that the physical and chemical

properties of red mud depend on the bauxite used and the manner in which the bauxite is

processed. Detailed characterization of red mud generated from NALCO aluminum refinery

at Damanjodi, India is reported by Mohapatra et al. [2] and of some other sources by various

authors [3]. Till today, almost all over the world, red mud is disposed off the plant site in two

main ways depending on the facilities available and the surroundings. In countries such as

France, England, Germany or Japan where availability of land for dumping is less and sea is

nearby; the practice is to discharge the mud into the sea. Where free land is available nearby,

the mud is pumped into pools and ponds constructed for this purpose. Probably the easiest

use for the mud is some sort of useful landfill instead of just dumping. Some attempts in this

direction are: filling material for mined or quarrying areas, land fill cover, road bed and levee

material, alternative to natural marsh sediment, agricultural land soil neutralization,

composting domestic waste etc. For use in many of these areas some sort of neutralization or

red mud amendment becomes necessary. A lot of efforts are being made globally to find out

suitable uses of red mud so that alumina industry may end up with no residue at all [4]. For

complete utilization of red mud, Thakur et al. [5] and Kovalenko [6] proposed avenues such

as building material production as an additive to cement, production of colouring agent for

paint works for ground floors of industrial and other buildings, of toned paper in the wood-

pulp and paper industry, of iron ore sinter and pellets in the ferrous metallurgy and in

agriculture for the purpose of improvement of the soil structure and as a micro fertilizer and a

neutralizer of pesticides. But a possibility that the incorporation of red mud particles along

with synthetic fibers in polymer could provide a synergism in terms of improved performance

such as better wear resistance has not been addressed so far.

Against this background, the present research work has been undertaken, with an

objective to explore the potential of red mud as a filler material in polymer composites and to

investigate its effect on the erosion wear performance of the resulting composites. Red mud is

accumulated at the alumina plant sites at an increasing rate throughout the world (nearly 30

million tons per year) and this work is an attempt to find a possible use of this abundant waste

which might gainfully be employed as particulate filler in polymers for developing low cost,

light weight, high strength and erosion wear resistant composites.

Erosion wear is caused by the impact of particles of solid or liquid against the surface

of an object. It occurs in a wide variety of machinery and typical examples are the damage to

gas turbine blades when an aircraft flies through dust clouds and the wear of pump impellers

in mineral slurry processing systems. In common with other forms of wear, mechanical

strength does not guarantee wear resistance and a detailed study of material characteristics is

required for wear minimization. The properties of the eroding particle are also significant and

are increasingly being recognized as a relevant parameter in the control of this type of wear.

This term ‘erosive wear’ in reality refers to an unspecified number of wear mechanisms

which occur when relatively small particles impact against mechanical components. This

definition is empirical by nature and relates more to practical considerations than to any

fundamental understanding of wear. Polymers are gaining importance as erosive wear

resistant materials for engineering and structural applications such as aerospace, automobile,

shipbuilding and other industries. As a result, much attention is focused on the study of solid

particle erosion behaviour of polymer composites. Erosion resistance has thus become an

important material property, particularly in the selection of alternative materials. There are

several research papers available in the literature that discuss about the erosive wear

behaviour of fiber reinforced composites [7, 8]. These investigations are mainly focused on

the study of influence of experimental and target related parameters on erosion wear rate.

Ceramic fillers have been used with different polymer matrices such as polypropylene

and nylon to study relation between the mechanical properties (tension and compression)

with particle size and particle volume fraction [9]. For designing of proper composites

satisfying various functional requirements, a large number of criteria must be satisfied

simultaneously. The filler content largely influences the density, tensile strength and wear

characteristics of the composites. While the density of the composite increases with the

increase in filler content, the tensile strength may decrease. Again, the wear characteristics of

the composites are not only dictated by filler content but also by the operating conditions. In

such situation, it is really a challenging task for the composite engineers to design a proper

composite satisfying all functional requirements. These engineering composites are expected

to have characteristics like ease of fabrication, low cost and high corrosion resistance. It

should also possess desirable properties such low density, high tensile strength and high wear

resistance [10]. In order to study and design the composite meeting the multiple desirable

performance criteria, a methodology must be evolved. To this end, an attempt has been made

in this study to analyze the impact of more than one parameter and their interactions using

Taguchi’s parameter design on the erosion wear of the red mud filled epoxy-glass fiber

composites. Such an approach has been successfully applied for parametric appraisal in the

wire electrical discharge machining (WEDM) process, drilling of metal matrix composites

and erosion behaviour of polymer–matrix composites [8, 11-15].

As already mentioned, to obtain the desired properties from a hybrid composite

system, reinforcement and fillers are added to the polymers. The additional improvements in

mechanical and tribological properties are in many cases attained through the incorporation

of glass or carbon fiber reinforcement and through the filling of particulate matters.

However, tribo properties are not intrinsic material properties, but strongly depend

upon the system in which material functions. So the influence of fillers and fibers on the

tribo-behaviour of composites cannot be predicted a priori and has to be tested in the

laboratory. In many industrial applications of composites, an understanding of tribological

behaviour is also necessary along with an understanding of the mechanical properties. Hence,

the primary concern in the present study has been to study how the red mud filled glass-

epoxy composites respond to the impact of erodent particles under different operating

conditions, to assess the damage due to wear and finally to determine the optimal parameter

settings for minimum wear loss. A mathematical model has been developed to determine the

erosion rate as a function of process and material variables so that results of theoretical and

experimental data can be compared to gain insight into the wear mechanism. Furthermore,

the analyses of variance are employed to investigate the most significant control factors and

their interactions.

2. Mathematical model

Nomenclature

a characteristic dimension of the square pyramidal shaped erodent i.e. the height and base length (m)

δ indentation depth (m) ev volumetric wear loss per particle impact (m3)

EV total volumetric erosion wear rate (m3/sec) α angle of impingement (degree) U impact velocity (m/sec) P force on the indenter (N) Hv hardness (N/m2) m mass of single erodent particle (kg) M mass flow rate of the erodent (kg/sec) N number of impact per unit time (sec-1) θ erodent temperature (0C) θ0 room temperature (0C) ρc density of composite (kg/m3) ρ density of erodent (kg/m3) ηnormal erosion efficiency with normal impact η erosion efficiency S specific heat of silica sand (J/Kg K) Er erosion rate (kg/kg) Erth theoretical erosion wear rate (kg/kg)

Solid particle erosion is a wear process in which the material is removed from a

surface by the action of a high velocity stream of erodent particles entrained in a high

velocity fluid stream. The particles strike against the surface and promote material loss.

During flight, a particle carries momentum and kinetic energy which can be dissipated during

the impact due to its interaction with a target surface. As far as erosion study of polymer

matrix composites is concerned, no specific model has been developed and thus the study of

their erosion behaviour has been mostly experimental. However, Mishra [16] proposed a

mathematical model for material removal rate in abrasive jet machining process in which the

material is removed from the work piece in a similar fashion. This model assumes that the

volume of material removed is same as the volume of indentation caused by the impact. This

has a serious limitation as in a real erosion process the volume of material removed is

actually different from the indentation volume. Further, this model considers only the normal

impact i.e α = 900 whereas in actual practice, particles may impinge on the surface at any

angle ( 00 900 ≤≤ α ). The proposed model addresses these shortcomings in an effective

manner. It considers the real situation in which the volume of material removed by erosion is

not same as the volume of material displaced and therefore, an additional term “erosion

efficiency (η)” is incorporated in the erosion wear rate formulation. In the case of a stream of

particles impacting a surface normally (i.e. at α = 900), erosion efficiency (ηnormal) defined by

Sundararajan et al. [17] is given as

2Uρ

ErHv2normalη = (1)

But considering impact of erodent at any angle α to the surface, the actual erosion

efficiency can be obtained by modifying Eq. (1) as

α2Sin2Uρ

ErHv2η = (2)

Another model proposed recently by Patnaik et al. [8,12] assumes that the kinetic

energy of the impinging particles is utilized to cause indentation in the composite surface and

the material loss is a measure of this indentation. It also assumes that both the erodent

material and the target material are at same temperature and therefore there is no exchange of

any thermal energy between them during the impact. This may be true for a room temperature

erosion situation, but when the erodent is at an elevated temperature, as in the case of hot air

carrying pulverised coal powders in a pipe, there will be dissipation of the kinetic energy as

well as the thermal energy from the erodent body to the target. Research on erosion of

composite materials by high temperature erodent particles is rare and there is no specific

model that includes the possibility of this thermal energy contributing to the magnitude of

erosion wear. Besides, while all previous models have been developed assuming the shape of

erodent to be spherical, in the real situation, the erodent particles are actually irregular shaped

bodies having sharp edges (Fig. 1). Considering them to be square pyramidal shaped bodies is

a more realistic assumption as compared to assuming them simply spherical. The model

proposed in the present work addresses to all these shortcomings. It assumes the erodent

particles to be rigid, square pyramidal shaped bodies of height and base length equal to the

average grit size. It is further based on the assumption that the loss in both kinetic as well as

thermal energy of the impinging particles is utilized to cause micro-indentation in the

composite material and the material loss is a measure of the indentation. The erosion is the

result of cumulative damage of such non-interacting, single particle impacts. The model is

developed with the simplified approach of energy conservation which equals the loss in

erodent kinetic energy and thermal energy during impact with the work done in creating the

indentation. It proceeds as follows.

At time t after initial contact, the particle of mass m will have indented the surface to

a depth x; the cross-sectional area of the indentation at the surface will be A(x), where A(x)

normally determined by the shape of the erodent particle. The material removal mechanism

has been schematically shown in Fig. 2. The upward force decelerating the particle will be

that due to the plastic flow pressure acting over A (x); and the equation of motion of the

particle can therefore be written as:

HA(x)2dt

x2dm −= (3)

For simple particle shapes, this equation can readily be solved analytically. But to

know the final volume of indentation when the particle comes to rest at a depth δ at time t=

T, as shown in Fig. 2, the work done by the retarding force will equal to the sum of the

kinetic energy and the loss of thermal energy of the particle.

The conservation of energy can be represented by the equation

)θS( θ.mU.m21dxA(x)H 0

2δ

0

−+=∫ (4)

The impact velocity will have two components; one normal to the composite surface

and one parallel to it. At zero impact angles, it is assumed that there is negligible wear

because eroding particles do not practically impact the target surface [18]. Consequently,

there will be no erosion due to the parallel component and the indentation is assumed to be

caused entirely by the component normal to the composite surface as shown in Figure 3.

Thus, in case of oblique impact, the kinetic energy corresponding to the normal component of

velocity is considered and Eq. (4) becomes:

)θS(θ.mαSinU.m21dxHA(x) 0

22δ

0

−+=∫ (5)

Now,

3

3δdxδ

02xdx

δ

0A(x) =∫=∫

So, the volumetric wear loss per particle impact is given by

ev = Volume of indentation×η

= 3δ.η

3

Considering N number of particle impacts per unit time, the volumetric erosion wear loss will

be

η3δNEv

3

=

Now applying conservation of energy to the single impact erosion process, the sum of

the kinetic energy associated with the normal velocity component and the loss of thermal

energy of the a single erodent particle is equal to the work done in the indentation of

composite. The energy of impact introduces a force P on the indenter to cause the indentation

in the composite. Thus,

=× δP.21 )θS(θ.mαSin.U.m

21

022 −+

2

)θS(θ.m2.αSin.U.mHδ21 0

223 −+

=×

⎥⎥

⎦

⎤

⎢⎢

⎣

⎡ −+=

H3

)0θS(θ.m2α2Sin.2U.m.ηve

For multiple impact

⎥⎦

⎤⎢⎣

⎡ −+=

H3)θ.(θS2.αSin.UN.m.ηE 0

22

V

Or, ⎥⎦

⎤⎢⎣

⎡ −+=

H3)θ.(θS2.αSin.UM.ηE 0

22

V

The non-dimensional erosion rate, defined as the composite mass lost per unit time

due to erosion divided by the mass of the erodent causing the loss, is now expressed as

[ ])θS(θ2.αSinUH3ρ.ηE 0

22Cr −+= (6)

The mathematical expression in Eq. (6) can be used for predictive purpose to make an

approximate assessment of the erosion damage from the composite surface. When the

erodent temperature is same as room temperature, Eq.(6) reduces to:

[ ]α2Sin2UH3

cρ.ηrE = (7)

Here in Eq.(7), the role of thermal energy transfer from erodent to target material in

causing erosion is absent and thus the expression is similar to the one in the theoretical model

proposed earlier by Patnaik et al.[8].

Material removal by impact erosion wear involves complex mechanisms. A simplified

theoretical model for such a process may appear inadequate unless its assessment against

experimental results is made. So for the validation of the proposed model erosion tests on the

composites are conducted at various operating conditions.

3. Experimental details

3.1. Composite fabrication

E-glass fibers (360 roving taken from Saint Gobian Ltd.) are reinforced in red mud

filled Epoxy LY 556, chemically belonging to the ‘epoxide’ family is used as the matrix

material. Its common name is Bisphenol-A-Diglycidyl-Ether. The low temperature curing

epoxy resin (LY 556) and corresponding hardener (HY951) are mixed in a ratio of 10:1 by

weight as recommended. The epoxy resin and the hardener are supplied by Ciba Geigy India

Ltd. Red mud collected from NALCO aluminium refinery at Damanjodi, India is sieved to

obtain particle size in the range 70-90 μm. E-glass fiber and epoxy resin have modulus of

72.5GPa and 3.42GPa respectively and possess density of 2590 kg/m3 and 1100 kg/m3

respectively. Composites of three different compositions (0wt%, 10wt% and 20wt% red mud

filling) are made and the fiber loading (weight fraction of glass fiber in the composite) is kept

at 50% for all the samples. The castings are put under load for about 24 hours for proper

curing at room temperature. Specimens of suitable dimension are cut using a diamond cutter

for physical characterization and erosion test.

3.2. Test of micro-hardness, density, tensile and flexural properties

Density

The theoretical density of composite materials in terms of weight fraction can easily

be obtained as for the following equations given by Agarwal and Broutman [19].

( ) ( )mmffct WW ρρ

ρ//

1+

= (8)

Where, W and ρ represent the weight fraction and density respectively. The suffix f, m and ct

stand for the fiber, matrix and the composite materials respectively.

The composites under this investigation consists of three components namely matrix,

fiber and particulate filler. Hence the modified form of the expression for the density of the

composite can be written as

( ) ( ) ( )ppmmffct WWW ρρρ

ρ///

1++

= (9)

Where, the suffix ‘p’ indicates the particulate filler materials.

The actual density ( ceρ ) of the composite, however, can be determined

experimentally by simple water immersion technique. The volume fraction of voids ( vV ) in

the composites is calculated using the following equation:

ct

cectvV

ρρρ −

= (10)

Micro-hardness

Micro-hardness measurement is done using a Leitz micro-hardness tester. A diamond

indenter, in the form of a right pyramid with a square base and an angle 1360 between

opposite faces, is forced into the material under a load F. The two diagonals X and Y of the

indentation left on the surface of the material after removal of the load are measured and their

arithmetic mean L is calculated. In the present study, the load considered F = 24.54N and

Vickers hardness number is calculated using the following equation.

21889.0LFHV = (11)

and 2

YXL +=

Where F is the applied load (N), L is the diagonal of square impression (mm), X is the

horizontal length (mm) and Y is the vertical length (mm).

Tensile strength

The tensile test is generally performed on flat specimens. The commonly used

specimen for tensile test is the dog-bone specimen and straight side specimen with end tabs.

A uniaxial load is applied through both the ends. The ASTM standard test method for tensile

properties of fiber resin composites has the designation D 3039-76. The length of the test

section should be 200 mm. The tensile test is performed in the universal testing machine

Instron 1195 and results are analyzed to calculate the tensile strength of composite samples.

Flexural and Inter-laminar shear strength

The flexural strength of a composite is the maximum tensile stress that it can

withstand during bending before reaching the breaking point. The three point bend test is

conducted on all the composite samples in the universal testing machine Instron 1195. Span

length of 40 mm and the cross head speed of 10mm/min are maintained.

Impact strength

Low velocity instrumented impact tests are carried out on composite specimens. The

tests are done as per ASTM D 256 using an impact tester. The pendulum impact testing

machine ascertains the notch impact strength of the material by shattering the V-notched

specimen with a pendulum hammer, measuring the spent energy, and relating it to the cross

section of the specimen. The standard specimen for ASTM D 256 is 64 x 12.7 x 3.2 mm3 and

the depth under the notch is 10.2 mm. The machine is adjusted such that the blade on the

free-hanging pendulum just barely contracts the specimen (zero position). Since there are

practically no losses due to bearing friction, etc. (< 0.3 %), the testing conditions may be

regarded as ideal. The specimens are clamped in a square support and are struck at their

central point by a hemispherical bolt of diameter 5 mm. The respective values of impact

energy of different specimens are recorded directly from the dial indicator.

3.3. Erosion test apparatus

The solid particle erosion experiments are carried out as per ASTM G76 on the

erosion test rig shown schematically in Fig. 4. The test rig consists of an air compressor, an

air drying unit, a conveyor belt-type particle feeder and an air particle mixing and

accelerating chamber. The dried and compressed air is then mixed with the silica sand (300–

600µm size) which is fed constantly by a conveyor belt feeder into the mixing chamber and

then accelerated by passing the mixture through a convergent brass nozzle of 3 mm internal

diameter. The set up is capable of creating reproducible erosive situations for assessing

erosion wear resistance of the composite samples. The erodent particles impact the specimen

which can be held at different angles with respect to the direction of erodent flow using a

swivel and an adjustable sample clip. The velocity of the eroding particles is determined

using standard double disc method [20]. The apparatus is equipped with a heater which can

regulate and maintain the erodent temperature at any pre-determined fixed value during an

erosion trial. In the present study, dry silica sand (assumed to be square pyramidal shaped) of

different particle sizes (300µm, 450µm and 600µm) are used as erodent. The samples are

cleaned in acetone, dried and weighed to an accuracy of ± 0.1 mg before and after the erosion

trials using a precision electronic balance. The weight loss is recorded for subsequent

calculation of erosion rate. The process is repeated till the erosion rate attains a constant value

called steady state erosion rate.

3.4. Experimental design

Design of experiment is a powerful analysis tool for modelling and analyzing the

influence of control factors on performance output. The most important stage in the design of

experiment lies in the selection of the control factors. Therefore, a large number of factors are

included so that non-significant variables can be identified at earliest opportunity. Exhaustive

literature review on erosion behaviour of polymer composites reveal that parameters viz.,

impact velocity, filler content, erodent temperature, impingement angle, stand-off distance

and erodent size etc largely influence the erosion rate of polymer composites [8,11,17]. The

impact of six such parameters are studied using L27 (313) orthogonal design. The fixed and

variable parameters chosen for the test are given in Table 1. The selected levels of the six

control parameters are listed in Table 2.

In Table 3, each column represents a test parameter whereas a row stands for a

treatment or test condition which is nothing but combination of parameter levels. In

conventional full factorial experiment design, it would require 36 = 729 runs to study five

parameters each at three levels whereas, Taguchi’s factorial experiment approach reduces it

to only 27 runs offering a great advantage in terms of experimental time and cost. The

experimental observations are further transformed into signal-to-noise (S/N) ratio. There are

several S/N ratios available depending on the type of performance characteristics. The S/N

ratio for minimum erosion rate can be expressed as “lower is better” characteristic, which is

calculated as logarithmic transformation of loss function as shown below.

Smaller is the better characteristic: ( )∑−= 21log10 ynN

S (12)

Where ‘n’ the number of observations and y the observed data. The standard linear

graph, as shown in Fig. 5, is used to assign the factors and interactions to various columns of

the orthogonal array [21]. The plan of the experiments is as follows: the first column of this

orthogonal array is assigned to impact velocity (A), the second column to red mud content

(B), the fifth column to erodent temperature (C), the ninth column to impingement angle (D),

the tenth column to stand-off distance (E) and the twelfth column to erodent size (F), the third

and fourth column are assigned to (A×B)1 and (A×B)2 respectively to estimate interaction

between impact velocity (A) and red mud content (B),the sixth and seventh column are

assigned to (B×C)1 and (B×C)2 respectively to estimate interaction between red mud content

(B) and erodent temperature (C), the eight and eleventh column are assigned to (A×C)1 and

(A×C)2 respectively to estimate interaction between the impact velocity (A) and erodent

temperature (C) and the remaining columns are used to estimate experimental errors.

3.5. Scanning electron microscopy

The surfaces of the specimens are examined directly by scanning electron microscope

(SEM) JEOL JSM-6480LV. The eroded samples are mounted on stubs with silver past. To

enhance the conductivity of the eroded samples, a thin film of platinum is vacuum-evaporated

onto them before the photomicrographs are taken.

4. Results and discussion

4.1. Mechanical properties

The present investigation reveals that the presence of red mud has varied effect on the

glass-epoxy composites in terms of mechanical properties. As is seen in Fig. 6a, the

composite micro-hardness has significantly improved with the addition of red mud. The

density too has increased with red mud filling. The neat epoxy taken for this study possess a

density of 1.1 gm/cc which increases to 1.53 gm/cc (with a void fraction of 0.9%) with the

reinforcement of 50 wt% of glass fiber in it. But when this glass reinforced epoxy is filled

with micro sized red mud particles, the density of the resulting hybrid composites assume

higher values i.e. 1.65 gm/cc (void fraction of 3.225%) and 1.752 gm/cc (void fraction of

7.894%) for composites with red mud contents of 10 and 20 wt% respectively.

Consequently, the void fraction in the composites too increases with the filler content. Fig. 6b

presents the tensile and flexural strengths of the composites with and without red mud filling.

Similarly, Fig. 6c illustrates the variation of tensile modulus with the weight fraction of red

mud in the composites. It can be seen that the tensile properties have become distinctly

poorer with the incorporation of red mud particles in the matrix. Previous reports [22,23]

demonstrate that normally the glass fibers in the composite restrain the deformation of the

matrix polymer reducing the tensile strain. So even if the strength decreases with filler

addition the tensile modulus of the hybrid composite is expected to increase. But this is

possibly not occurring in the present case with the presence of red mud as the filler and as a

result reduction in both tensile strength and modulus is recorded in spite of the reinforcement

of long glass fibers. This might have influenced the flexural strength of these hybrid

composites as well which is showing (Fig. 6b) a decreasing trend up to a filler content of 10

wt%.

Short beam shear test is carried out on these laminates to determine the inter-laminar

shear strength (ILSS) of composites prepared with different filler content. Apparent ILSS of

glass-fiber reinforced epoxy resin is measured as 99.75MPa. With the addition of red mud,

ILSS of laminates decreases drastically as is seen in case of the composite with filler up to

10wt% (Fig. 6d). The reduction may be related with the formation of voids in the matrix

which is generally located at the inter-laminar region of composites. However, for the

composite with 20wt% filler, a relatively higher value of ILSS has been recorded. Impact

strength is the ability of a material to resist the fracture under stress applied at high speed.

The impact properties of composite materials are directly related to its overall toughness.

Composite fracture toughness is affected by inter-laminar and inter-facial strength

parameters. Fig. 6e shows the variation of work of fracture (impact strength) of glass-epoxy

composites with and without red mud filling. The presence of red mud is seen to have

improved the impact strength of these composites.

4.2. Surface morphology

Fig. 7 presents the SEM of the unfilled glass-epoxy composite surfaces eroded under

various test conditions. In Fig. 7 (a) no cracks or craters are seen on the composite surface

after erosion due to impact of dry silica sand particles (temperature 400C) of smallest grit size

(300 µm) with a lower impact velocity (43 m/sec) at a low impingement angle of 300. But as

the erosion tests are carried out with higher erodent temperature (600C) and grit size (450

µm), the morphology of the eroded surface becomes different as in Fig. 7(b). The matrix is

chipped off and the glass fibers are clearly visible beneath the matrix layer. The micrographs

with higher magnifications presented in Figs. 7 (c), (d) and (e) distinctly illustrate the craters

formed due to material loss and the arrays of broken and semi-broken glass fibers within. Due

to repeated impact of hard and high temperature sand particles there is initiation of cracks on

the fibers and as erosion progresses, these cracks subsequently propagate on the fiber bodies

both in transverse as well as longitudinal manner. Such cracks are clearly noticed in Figs.

7(f), (g) and (h) respectively. It is also evident from the microstructures of these eroded

surfaces that for higher erodent temperature and impact velocity, the damage to the surface is

also relatively greater. From SEM observations of the eroded surfaces, shown in Fig. 8, of the

glass-epoxy composites filled with different weight proportions of red mud, it appears that

composites under consideration exhibit several stages of erosion and material removal

process Very small craters and short cracks are seen on the eroded surface of the composite

with red mud (10 wt%), in Fig. 8(a), indicating the initiation of matrix material loss from the

surface. Figs. 8(a) and (b) also show signs of plastic deformation of the matrix material and

when impacting at such a low angle (300), the hard erodent particles penetrate the surface and

cause material removal mostly by micro-ploughing. Fig. 8(c) shows the micrograph of the

same composite surface eroded at an impingement angle of 600 and an impact velocity of 54

m/sec. The matrix covering the fiber seems to be chipped off and the crater thus formed

shows the fiber body which is almost intact. Repeated impact of the erodent has caused

roughening of the surface. Fragmentation of the fibers as a result of cracks and multiple

fractures are also distinctly shown in micrograph given in Fig. 8(d). Figs. 8 (e), (f), (g) and

(h) are the SEM images for the glass-epoxy composites filled with 20% red mud. After the

local removal of matrix, the arrays of fibers are normally exposed to erosive environment. At

low impact velocity and impingement angle, the damage to the surface is minimal as seen in

Figs. 8(e) and (f). Subsequently the material removal becomes faster. The wear trace is

distinctly visible, and there is protrusion of fibers beneath the matrix layer as seen in Fig.

8(g). The broken fiber and red mud filler fragments, seen in Fig. 8(h), are mixed with the

matrix micro-flake debris and the damage of the composite is characterized by separation and

detachment of this debris

Particle impingement produces a rise in temperature of the surface which makes the

matrix deformation easy because the high temperature known to occur in solid particle

erosion invariably soften the matrix. On impact the erodent particle kinetic energy is

transferred to the composite body that leads to crater formation and subsequently material

loss. The presence of hard red mud particle in the matrix helps in absorbing a good fraction of

this kinetic energy and therefore energy available for the plastic deformation of epoxy

becomes less. This also delays the initiation of fiber exposure as compared to the composite

without any filler. All these factors combined together result in exhibition of better erosion

response by the red mud filled composites than that of fiber reinforced epoxy without

particulate filling.

4.3. Steady state erosion

The erosion wear behaviour of polymer composites can be grouped into ductile and brittle

categories although this grouping is not definitive because the erosion characteristics depend

on the experimental conditions as much as on composition of the target material. It is well

known that impingement angle is one of the most important parameters in the erosion process

and for ductile materials the peak erosion normally occurs at 150 to 200 angle while for brittle

materials the erosion damage is maximum usually at normal impact i.e. at 900 impingement

angle. In the present study, the variation of erosion wear rate of the composites with

impingement angle is studied by conducting experiments under specified operating

conditions. The results are presented in Fig. 9 which shows the peak erosion taking place at

an impingement angle of 600 for the unfilled as well as the red mud filled glass-epoxy

composites. This clearly indicates that these composites respond to solid particle impact

neither in a purely ductile nor in a purely brittle manner. This behaviour can be termed as

semi-ductile in nature which may be attributed to the incorporation of glass fibres and red

mud particles within the epoxy body. Similarly, the variation of erosion rate of unfilled and

red mud filled composites with erodent temperature is shown in Figure 10. Erosion trials are

conducted at seven different temperatures under normal impact condition. It is seen, in this

figure, that for all the composite samples, the erosion rates remain almost unaffected by the

change in erodent temperature from ambient to 400C. The effect of erodent temperature on

erosion is significant above 40 0C and the rate of increase in erosion rate is greater at higher

temperatures. The increase in erosion rate with erodent temperature can be attributed to

increased penetration of particles on impact as a result of dissipation of greater amount of

particle thermal energy to the target surface. This leads to more surface damage, enhanced

sub-critical crack growth etc. and consequently to the reduction in erosion resistance.

In Table 3, the last column represents S/N ratio of the erosion rate which is in fact the

average of two replications. The overall mean for the S/N ratio of the erosion rate is found to

be -47.47 db. The analysis was made using the popular software specifically used for design

of experiment applications known as MINITAB 14. Before any attempt is made to use this

simple model as a predictor for the measure of performance, the possible interactions

between the control factors must be considered. Thus factorial design incorporates a simple

means of testing for the presence of the interaction effects. Analysis of the result leads to the

conclusion that factor combination of A2, B2, C1, D1, E2 and F3 gives minimum erosion rate

(Fig. 11). The interaction graphs are shown in the Figs. 12a, 12b and 12c.

4.4. Erosion efficiency

The hardness alone is unable to provide sufficient correlation with erosion rate,

largely because it determines only the volume displaced by each impact and not really the

volume eroded. Thus a parameter which will reflect the efficiency with which the volume

that is displaced is removed should be combined with hardness to obtain a better correlation.

The erosion efficiency is obviously one such parameter. This thought has already been

reflected in the theoretical model but the evaluation of erosion efficiency can be made only

on the basis of experimental data. Hence, the values of erosion efficiencies of these

composites calculated using Eq. (2) is summarized in Table 4 along with their hardness

values and operating conditions. It clearly shows that erosion efficiency is not exclusively a

material property; but also depends on other operational variables such as impingement angle

and impact velocity. The erosion efficiencies of these composites under normal impact

(ηnormal) vary from 3 to 6%, 6-9% and 9-12% for impact velocities 65m/sec, 54m/sec and

43m/sec respectively. The value of η for a particular impact velocity under oblique impact

can be obtained simply by multiplying a factor 1/Sin2α with ηnormal. Similar observation on

velocity dependence of erosion efficiency has previously been reported by few investigators

[24].

The theoretical erosion wear rate (Erth) of the red mud filled glass-epoxy composites are

calculated using Eq. 7. These values are compared with those obtained from experiments

(Erexpt) conducted under similar operating conditions. Table 5 presents a comparison among

the theoretical and experimental results and the corresponding associated error percentage.

The errors in experimental results with respect to the theoretical ones lie in the range 0-10%.

The magnitude of η can be used to characterize the nature and mechanism of erosion. For

example, ideal micro-ploughing involving just the displacement of the material from the

crater without any fracture (and hence no erosion) will results in η=0. In contrast, if the

material removal is by ideal micro-cutting, η = 1.0 or 100%. If erosion occurs by lip or

platelet formation and their fracture by repeated impact, as is usually the case in the case of

ductile materials, the magnitude of η will be very low, i.e η ≤ 100%. In the case of brittle

materials, erosion occurs usually by spalling and removal of large chunks of materials

resulting from the interlinking of lateral or radial cracks and thus η can be expected to be

even greater than 100% [25]. The erosion efficiencies of the composites under the present

study indicate that at low impact speed the erosion response is semi-ductile (η=10-100%). On

the other hand at relatively higher impact velocity the composites exhibit ductile (η < 10%)

erosion behavior [24].

4.5. ANOVA and the effects of factors

In order to find out statistical significance of various factors like impact velocity (A),

red mud content (B), erodent temperature (C), impingement angle (D), stand-off distance (E)

and erodent size (F) on erosion rate, analysis of variance (ANOVA) is performed on

experimental data. Table 6 shows the results of the ANOVA with the erosion rate. This

analysis is undertaken for a level of confidence of significance of 5 %. The last column of the

table indicates that the main effects are highly significant (all have very small p-values).

From Table 6, one can observe that red mud content (p=0.009), erodent temperature

(p = 0.018), stand-off distance (p = 0.026), impingement angle (p= 0.053) and impact

velocity (p=0.084) have great influence on erosion rate. The interaction of impact

velocity× red mud content (p=0.018) and erodent temperature× red mud content (p=0.027)

show significant contribution on the erosion rate but the remaining factors and interactions

have relatively less significant contribution on erosion rate.

5. Confirmation experiment

The combination of control factors has been determined in the previous analysis.

However, the final step in any design of experiment approach is to predict and verify

improvements in observed values through the use of the optimal combination level of control

factors. The confirmation experiment is performed by taking an arbitrary set of factor

combination A3B3C2D3E1, but factor F has been omitted and factor A has also least effect on

performance characteristics, but factor A with factor B interaction have significant effect on

minimum erosion rate as evident from Table 6. Therefore, factor A cannot be omitted from

this series. The estimated S/N ratio for erosion rate can be calculated with the help of

following prediction equation:

( ) )10(TE)TD()]TC()TB(T)CB[()TC()]TB()TA()TBA[()TB()TA(Tη̂

1323

2323333331

−+−+−−−−

−+−+−−−−−+−+−+=

1η Predicted average

T Overall experimental average

13233 EandD,C,B,A Mean response for factors and interactions at designated levels.

By combining like terms, the equation reduces to

T2EDBCBBAη 13323331 −++−+= (11)

A new combination of factor levels A3, B3, C2, D3 and E1 is used to predict deposition rate

through prediction equation and it is found to be dB1η 47.8035-= .

For each performance measure, an experiment is conducted for a different factors

combination and compared with the result obtained from the predictive equation as shown in

Table 7. The resulting model seems to be capable of predicting erosion rate to a reasonable

accuracy. An error of 3.61 % for the S/N ratio of erosion rate is observed. However, the error

can be further reduced if the number of measurements is increased. This validates the

development of the mathematical model for predicting the measures of performance based on

knowledge of the input parameters.

5. Conclusions

This analytical and experimental investigation into the erosion behaviour of red mud filled

glass-epoxy hybrid composites leads to the following conclusions:

1. Hybrid composites suitable for applications in highly erosive environments can be

prepared by reinforcement of glass fibres and filling of micro-sized red mud particles

in epoxy resin. The erosion wear performance of these composites improves quite

significantly by addition of red mud filler.

2. A mathematical model based on conservation of particle kinetic energy during

multiple impact erosion process has been developed. To overcome the shortcomings

of the existing theoretical models an ‘erosion efficiency’ term has been introduced. It

is demonstrated that if supported by an appropriate magnitude of erosion efficiency,

the model can perform well for polymer based hybrid composites for normal as well

as oblique impacts.

3. Erosion characteristics of these composites can be successfully analyzed using

Taguchi experimental design scheme. Taguchi method provides a simple, systematic

and efficient methodology for the optimization of the control factors. Factors like red

mud content, erodent temperature, stand-off distance, impingement angle and impact

velocity in order of priority are significant to minimize the erosion rate. Although the

effect of impact velocity is less compared to other factors, it cannot be ignored

because it shows significant interaction with another factor i.e. the weight percentage

of red mud in the composite.

4. Study of influence of impingement angle on erosion rate of the composites filled with

different weight percentage of red mud reveals their semi-ductile nature with respect

to erosion wear. The peak erosion rate is found to be occurring at 600 impingement

angle for all the composite samples under various experimental conditions. The

erosion rate is also greatly affected by the erodent temperature.

5. Possible use of these composites in components such as pipes carrying coal dust,

industrial fans, desert structures, low cost housing, fishing boats / water-sports

equipments, false ceiling, partition boards etc. is recommended. In future, this study

can be extended to new hybrid composites using other potential fillers and the

resulting experimental findings can be similarly analyzed.

References [1]. Katz H S, Mileski J V. Handbook of Fillers for Plastics, November 30, A Von Nostrand

Reinhold Book; 1987. [2]. Mahapatra B K, Rao M B S, Bhima Rao R, Paul A K. Characteristics of Red Mud

Generated at NALCO Refinery, Damanjodi, India. Light Metals 2000; 161–165. [3]. Solimar K, Sajo I, Steiner J, Zoldi J. Characteristics and separability of red mud. LIGHT

METALS 1992; 209 - 223. [4]. Banvolgyi G, Siklosi P. The improved low temperature digestion (ILTD) process: An

economic and environmentally sustainable way of processing gibbsitic bauxites. LIGHT METALS 1998; 45 – 53.

[5]. Thakur R S, Das S N. Red Mud – Analysis and Utilization -- Publication and Information Directorate, New Delhi and Wiley Eastern Limited, New Delhi; 1994.

[6]. Kovalenko E P. Improvement of the process of alumina production at Nilolaev alumina plant, LIGHT METALS 1998; 55 – 58.

[7]. Harsha A P, Tewari US, Venkatraman B. Solid particle erosion behaviour of various polyaryletherketone composites. Wear 2003; 254:693-712.

[8]. Amar Patnaik, Alok Satapathy, Mahapatra SS, Dash R R. A Modeling Approach for Prediction of Erosion Behavior of Glass Fiber- Polyester Composites. Journal of Polymer Research doi: 10.1007/s10965-007-9154-2.

[9]. Wong K W Y, Truss R W. Effect of Flyash Content and Coupling Agent on the Mechanical Properties of Flyash-filled Polypropylene. Composites Science and Technology 1994; 52(3):361–368.

[10]. Zhu K, Schmauder S. Prediction of the Failure Properties of Short Fiber Reinforced Composites with Metal and Polymer Matrix. Computational Materials Science 2003; 28: 743–748.

[11]. Mahapatra S S, Patnaik Aamar. Optimization of Wire Electrical Discharge Machining (WEDM) Process Parameters using Genetic Algorithm. Int. J. Adv. Manuf. Technol. 2007; 34:911–925.

[12]. Amar Patnaik, Alok Satapathy, Mahapatra SS, Dash R R. Tribo-Performance of Polyester Hybrid Composites: Damage Assessment and Parameter Optimization using Taguchi Design. Materials and Design, doi:10.1016/j.matdes.2008.04.057.

[13]. Patnaik Amar, Alok Satapathy, S.S. Mahapatra and R.R.Dash. Erosive Wear Assesment of Glass Reinforced Polyester-Flyash Composites using Taguchi Method. International Polymer Processing 2008; 23(2): 1-8.

[14]. Mahapatra SS, Amar Patnaik, Alok Satapathy. Taguchi Method Applied to Parametric Appraisal of Erosion Behavior of GF-Reinforced Polyester Composites. Wear 2008; 265: 214–222.

[15]. Alok Satapathy, Amar Patnaik, Manoj Kumar Pradhan. A Study on Processing, Characterization and Erosion Behavior of Fish (Labeo-rohita) Scale Filled Epoxy Matrix Composites. Materials and Design doi:10.1016/j.matdes.2008.10.033.

[16]. Mishra P K. Nonconventional Machining, Narosa Publishing House, New Delhi; 1997. [17]. Sundararajan G, Roy M, Venkataraman B. Erosion Efficiency- A New parameter to

Characterize the Dominant Erosion Micro-mechanism. Wear 1990; 140: 369. [18]. Srivastava V K, Pawar A G. Solid Particle Erosion of Glass Fiber Reinforced Flyash

Filled Epoxy Resin Composites. Composite Science and Technology 2006; 66:3021–3028.

[19]. Agarwal B D, Broutman L J. Analysis and performance of fiber composites: Second Edition. John Wiley and Sons, Inc.; 1990.

[20]. Ruff A W, Ives L K. Measurement of solid particle velocity in erosive wear. Wear 1975; 35 (1): 195-199.

[21]. Glen S P. Taguchi Methods: A Hands on Approach, Addison-Wesley; 1993. [22]. Fu S-Y, Lauke B. Characterization of tensile behavior of hybrid short glass fiber/

calcite particles/ABS composites. Composites: A 1998; 29:575–83. [23]. Thomason J L, Vlug M A, Schipper G, Krikor HGLT. Influence of fibre length and

concentration on the properties of glass fibre reinforced polypropylene: Part 3. Strength and strain at failure. Composites: A 1996; 27:1075–1084.

[24]. Roy M, Vishwanathan B, Sundararajan G. Solid particle erosion of polymer matrix composites. Wear 1994; 171:149–161.

[25]. Suresh Arjula, Harsha A P. Study of erosion efficiency of polymers and polymer composites. Polymer testing 2006; 25:188-196.

List of tables

Table 1. Parameters of the setting Table 2. Levels for various control factors Table 3. Experimental design using L27 orthogonal array Table 4. Erosion efficiency of GF-reinforced red mud filled epoxy resin Table 5. Comparison of theoretical and experimental results Table 6. ANOVA table for erosion rate (Red mud filled composites) Table 7. Results of the confirmation experiments for Erosion rate

Table 1. Parameters of the setting

Control Factors Symbols Fixed parameters

Velocity of impact Factor A Erodent Silica sand

Fiber loading Factor B Erodent feed rate (g/min) 10.0± 1.0

Erodent Temperature Factor C Nozzle diameter (mm) 3

Impingement angle Factor D Length of nozzle (mm) 80

Stand-off distance Factor E

Erodent size Factor F

Table 2. Levels for various control factors

Control factor Level

I II III Units

A:Velocity of impact 43 54 65 m/sec

B:Filler content 0 10 20 %

C:Erodent Temperature 40 50 60 0C

D:Impingement angle 30 60 90 degree

E:Stand-off distance 65 75 85 mm

F:Erodent size 300 450 600 µm

Table 3. Experimental design using L27 orthogonal array

Expt. No.

A (m/sec)

B (%)

C (0C)

D (Degree)

E (mm)

F (µm)

Er

(mg/kg)

S/N ratio (db)

1 43 0 40 30 65 300 204.348 -46.2074

2 43 0 50 60 75 450 342.029 -50.6813

3 43 0 60 90 85 600 413.720 -52.3341

4 43 10 40 60 75 600 256.522 -48.1825

5 43 10 50 90 85 300 376.124 -51.5066

6 43 10 60 30 65 450 266.667 -48.5194

7 43 20 40 90 85 450 222.663 -46.953

8 43 20 50 30 65 600 121.739 -41.7086

9 43 20 60 60 75 300 175.362 -44.8787

10 54 0 40 60 85 450 226.087 -47.0855

11 54 0 50 90 65 600 353.623 -50.9708

12 54 0 60 30 75 300 382.147 -51.6446

13 54 10 40 90 65 300 139.130 -42.8684

14 54 10 50 30 75 450 157.342 -43.9369

15 54 10 60 60 85 600 191.304 -45.6345

16 54 20 40 30 75 600 140.192 -42.9345

17 54 20 50 60 85 300 274.638 -48.7752

18 54 20 60 90 65 450 226.087 -47.0855

19 65 0 40 90 75 600 163.768 -44.2846

20 65 0 50 30 85 300 359.420 -51.112

21 65 0 60 60 65 450 443.712 -52.942

22 65 10 40 30 85 450 173.913 -44.8066

23 65 10 50 60 65 600 198.193 -45.9418

24 65 10 60 90 75 300 168.116 -44.5122

25 65 20 40 60 65 300 318.152 -50.0527

26 65 20 50 90 75 450 214.493 -46.6283

27 65 20 60 30 85 600 295.652 -49.4156

Table 4. Erosion efficiency of GF-reinforced red mud filled epoxy resin

Expt. No.

Impact Velocity

(V) m/sec

Density of eroding material

(ρ) kg/m3

Hardness of eroding material

(Hv) MPa

Erosion rate (Er) mg/kg

Erosion efficiency

(η) 1 43 1530 24.80 204.348 14.0546 2 43 1530 24.80 342.029 7.84180 3 43 1530 24.80 413.720 7.11370 4 43 1650 37.05 256.522 8.14750 5 43 1650 37.05 376.124 8.95910 6 43 1650 37.05 266.667 25.4076 7 43 1752 43.05 222.663 5.80390 8 43 1752 43.05 121.739 12.6928 9 43 1752 43.05 175.362 6.09490 10 54 1530 24.80 226.087 3.28680 11 54 1530 24.80 353.623 3.85550 12 54 1530 24.80 382.147 16.6659 13 54 1650 37.05 139.130 2.10140 14 54 1650 37.05 157.342 9.50580 15 54 1650 37.05 191.304 3.85280 16 54 1752 43.05 140.192 9.26830 17 54 1752 43.05 274.638 6.05260 18 54 1752 43.05 226.087 3.73670 19 65 1530 24.80 163.768 1.23230 20 65 1530 24.80 359.420 10.8184 21 65 1530 24.80 443.712 4.45210 22 65 1650 37.05 173.913 5.03590 23 65 1650 37.05 198.193 2.75480 24 65 1650 37.05 168.116 1.75250 25 65 1752 43.05 318.152 4.83920 26 65 1752 43.05 214.493 2.44680 27 65 1752 43.05 295.652 13.4902

Table 5. Comparison of theoretical and experimental results

Table 6. ANOVA table for erosion rate (Red mud filled composites)

Source DF Seq SS Adj SS Adj MS F P A 2 6.633 6.633 3.316 10.97 0.084 B 2 67.430 67.430 33.715 111.52 0.009 C 2 33.668 33.668 16.834 55.68 0.018 D 2 10.717 10.717 5.358 17.72 0.053 E 2 22.225 22.225 11.112 36.76 0.026 F 2 6.069 6.069 3.034 10.04 0.091

A*B 4 67.335 67.335 16.834 55.68 0.018 A*C 4 7.254 7.254 1.813 6.00 0.148 B*C 4 43.687 43.687 10.922 36.13 0.027 Error 2 0.605 0.605 0.302 Total 26 265.621

Expt. No. Erth (mg/kg)

Erexpt. (mg/kg)

Error (%) (Erth -Erexpt)

1 231.712 204.348 11.8096 2 387.851 342.029 11.8144 3 479.506 413.720 13.7195 4 248.364 256.522 03.2847 5 346.136 376.124 08.6633 6 244.771 266.667 08.9455 7 254.253 222.663 12.4246 8 137.891 121.739 11.7138 9 203.632 175.362 13.8830

10 236.075 226.087 4.23090 11 329.054 353.623 07.4664 12 358.521 382.147 06.5898 13 132.813 139.130 04.7558 14 169.921 157.342 07.4032 15 218.533 191.304 12.4601 16 134.217 140.192 04.4512 17 264.174 274.638 03.9607 18 206.321 226.087 09.5801 19 159.040 163.768 02.9726 20 417.574 359.420 13.9267 21 475.002 443.712 6.58736 22 168.002 173.913 03.5180 23 205.786 198.193 03.6899 24 169.358 168.116 00.7335 25 307.745 318.152 03.3816 26 189.435 214.493 13.2274 27 310.147 295.652 04.6738

Table 7. Results of the confirmation experiments for Erosion rate

Optimal control parameters Prediction Experimental

Level A3 B3C2D3E1 A3 B3C2D3E1 S/N ratio for Erosion rate (db) -47.8035 -46.0775

List of Figures

Figure 1 Shape of the erodent used

Figure 2 Scheme of material removal mechanism

Figure 3 Resolution of impact velocity in normal and parallel directions

Figure 4 Schematic diagram of an erosion test rig

Figure 5 Linear graphs for L27 array

Figure 6 (b) Variation of tensile and flexural strength with red mud content

Figure 6 (c) Variation of tensile modulus of the composites with red mud content

Figure 6 (d) Variation of inter-laminar shear strength with red mud content

Figure 6 (e) Variation of impact strength of the composites with red mud content

Figure 7 SEM micrographs of the eroded surfaces of the unfilled glass-epoxy composites

Figure 8 SEM micrographs of the eroded glass-epoxy composites filled with red mud

Figure 9 Effect of impingement angle on the erosion wear rate of the composites

Figure 10 Effect of erodent temperature on the erosion wear rate of the composites

Figure 11 Effect of control factors on erosion rate

Figure 12a Interaction graph between A x B for erosion rate

Figure 12b Interaction graph between A x C for erosion rate.

Figure 12c Interaction graph between B x C for erosion rate.

Figure 1 Shape of the erodent used

Figure 2 Scheme of material removal mechanism

a

a

Figure 3 Resolution of impact velocity in normal and parallel directions

Figure 4 Schematic diagram of an erosion test rig

Figure 5 Linear graphs for L27 array

Figure 6 (a) Variation of micro-hardness of the composites with red mud content

Figure 6 (b) Variation of tensile and flexural strength with red mud content

A(1)

B(2)

C(5)

D(9) E(10) F(12) (13)

(3,4) (6,7)

(8,11)

Figure 6 (c) Variation of tensile modulus of the composites with red mud content

Figure 6 (d) Variation of inter-laminar shear strength with red mud content

Figure 6 (e) Variation of impact strength of the composites with red mud content

Figure 7 SEM micrographs of the eroded surfaces of the unfilled glass-epoxy composites

(a) (b)

(c) (d)

(e) (f)

(g)

(h)

Figure 8 SEM micrographs of the eroded glass-epoxy composites filled with red mud

(a) (b)

(c) (d)

(e) (f)

(g)

(h)

150

175

200

225

250

275

15 30 45 60 75 90

Impingement angle (degree)

Eros

ion

rate

(mg/

kg )

0 wt% red mud10 wt% red mud20 wt% red mud

Figure 9 Effect of impingement angle on the erosion wear rate of the composites

150

175

200

225

250

30 35 40 45 50 55 60

Erodent temperature ( deg. Celcius)

Eros

ion

rate

(mg/

kg )

0 wt% red mud10 wt% red mud20 wt% red mud

Figure 10 Effect of erodent temperature on the erosion wear rate of the composites

Mea

n of

SN

rati

os

655443

-46

-47

-48

-49

-5020100 605040

906030

-46

-47

-48

-49

-50857565 600450300

A B C

D E F

Main Effects Plot (data means) for SN ratios

Signal-to-noise: Smaller is better

Figure 11 Effect of control factors on erosion rate

A

SN r

atio

s

655443

-44

-45

-46

-47

-48

-49

-50

B

20

010

Interaction Plot (data means) for SN ratios

Signal-to-noise: Smaller is better

Figure 12a Interaction graph between A x B for erosion rate

A

SN r

atio

s

655443

-44

-45

-46

-47

-48

-49

C

60

4050

Interaction Plot (data means) for SN ratios

Signal-to-noise: Smaller is better

Figure 12b Interaction graph between A x C for erosion rate.

B

SN r

atio

s

20100

-45

-46

-47

-48

-49

-50

-51

-52

-53

C

60

4050

Interaction Plot (data means) for SN ratios

Signal-to-noise: Smaller is better

Figure 12c Interaction graph between B x C for erosion rate.